<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
<meta name="Generator" content="Microsoft Exchange Server">
<!-- converted from rtf -->
<style><!-- .EmailQuote { margin-left: 1pt; padding-left: 4pt; border-left: #800000 2px solid; } --></style>
</head>
<body>
<font face="Calibri" size="2"><span style="font-size:11pt;">
<div>Hi Evan,</div>
<div> </div>
<div>I noticed that in X86Subtarget constructor you set ‘<font face="Courier New" size="2"><span style="font-size:10pt;">X86SSELevel’</span></font><font face="Courier New" size="2"><span style="font-size:10pt;"> </span></font>member of the class as ‘<font face="Courier New" size="2"><span style="font-size:10pt;">NoMMXSSE</span></font>’
in the case when HasAVX member is set to ‘true’.</div>
<div>Effectively that invalidates SSE features for AVX architecture - because hasSSEn() accessors return ‘false’ when HasAVXn() is ‘true’.</div>
<div>I wonder whether this is the behavior which you’d like to enforce – as conceptually AVX architecture complements SSE rather than replaces it completely.</div>
<div> </div>
<div>I noticed this problem after discovering that LLVM fails to lower “sse2.fence” intrinsic when generating a code for AVX architecture – because this intrinsic is conditioned on hasSSE2() being ‘true’.</div>
<div>Is that case was somehow missed from regression testing, or there is another way to lower that intrinsic?</div>
<div> </div>
<div>I’d appreciate your clarifications.</div>
<div> </div>
<div>Best Regards,</div>
<div> Victor Umansky</div>
<div> </div>
</span></font>
<font face="monospace">---------------------------------------------------------------------<br>
Intel Israel (74) Limited<br>
<br>
This e-mail and any attachments may contain confidential material for<br>
the sole use of the intended recipient(s). Any review or distribution<br>
by others is strictly prohibited. If you are not the intended<br>
recipient, please contact the sender and delete all copies.</font></body>
</html>